Thomas Vandal
ABSTRACT
Deep Learning (DL) methods have been transforming computer vision with innovative adaptations to other domains including climate change. For DL to pervade Science and Engineering (S&EE) applications where risk management is a core component, well-characterized uncertainty estimates must accompany predictions. However, S&E observations and model-simulations often follow heavily skewed distributions and are not well modeled with DL approaches, since they usually optimize a Gaussian, or Euclidean, likelihood loss. Recent developments in Bayesian Deep Learning (BDL), which attempts to capture uncertainties from noisy observations, aleatoric, and from unknown model parameters, epistemic, provide us a foundation. Here we present a discrete-continuous BDL model with Gaussian and lognormal likelihoods for uncertainty quantification (UQ). We demonstrate the approach by developing UQ estimates on "DeepSD'', a super-resolution based DL model for Statistical Downscaling (SD) in climate applied to precipitation, which follows an extremely skewed distribution. We find that the discrete-continuous models outperform a basic Gaussian distribution in terms of predictive accuracy and uncertainty calibration. Furthermore, we find that the lognormal distribution, which can handle skewed distributions, produces quality uncertainty estimates at the extremes. Such results may be important across S&E, as well as other domains such as finance and economics, where extremes are often of significant interest. Furthermore, to our knowledge, this is the first UQ model in SD where both aleatoric and epistemic uncertainties are characterized.
- P. Baldi, P. Sadowski, and D. Whiteson. Searching for exotic particles in high-energy physics with deep learning. Nature communications, 5:4308, 2014.Google Scholar
Cross Ref
- D. Barber and C. M. Bishop. Ensemble learning in bayesian neural networks. NATO ASI SERIES F COMPUTER AND SYSTEMS SCIENCES, 168:215--238, 1998.Google Scholar
- S. Basu, S. Ganguly, S. Mukhopadhyay, R. DiBiano, M. Karki, and R. Nemani. Deepsat: a learning framework for satellite imagery. In Proceedings of the 23rd SIGSPATIAL International Conference on Advances in Geographic Information Systems, page 37. ACM, 2015. Google ScholarDigital Library
- G. Bürger, T. Murdock, A. Werner, S. Sobie, and A. Cannon. Downscaling extremes-an intercomparison of multiple statistical methods for present climate. Journal of Climate, 25(12):4366--4388, 2012.Google ScholarCross Ref
- G. Bürger, T. Q. Murdock, a. T. Werner, S. R. Sobie, and a. J. Cannon. Downscaling extremes-an intercomparison of multiple statistical methods for present climate. Journal of Climate, 25(12):4366--4388, June 2012.Google ScholarCross Ref
- A. J. Cannon and P. H. Whitfield. Downscaling recent streamflow conditions in british columbia, canada using ensemble neural network models. Journal of Hydrology, 259(1):136--151, 2002.Google ScholarCross Ref
- H.-K. Cho, K. P. Bowman, and G. R. North. A comparison of gamma and lognormal distributions for characterizing satellite rain rates from the tropical rainfall measuring mission. Journal of Applied Meteorology, 43(11):1586--1597, 2004.Google ScholarCross Ref
- C. Daly, M. Halbleib, J. I. Smith, W. P. Gibson, M. K. Doggett, G. H. Taylor, J. Curtis, and P. P. Pasteris. Physiographically sensitive mapping of climatological temperature and precipitation across the conterminous united states. International journal of climatology, 28(15):2031--2064, 2008.Google ScholarCross Ref
- C. Dong, C. C. Loy, K. He, and X. Tang. Learning a deep convolutional network for image super-resolution. In European Conference on Computer Vision, pages 184--199. Springer, 2014.Google ScholarCross Ref
- H. J. Fowler, S. Blenkinsop, and C. Tebaldi. Linking climate change modelling to impacts studies: recent advances in downscaling techniques for hydrological modelling. International journal of climatology, 27(12):1547--1578, 2007.Google ScholarCross Ref
- Y. Gal. Uncertainty in Deep Learning. PhD thesis, Ph. D. thesis, University of Cambridge, 2016.Google Scholar
- Y. Gal and Z. Ghahramani. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In international conference on machine learning, pages 1050--1059, 2016. Google ScholarDigital Library
- Y. Gal, J. Hron, and A. Kendall. Concrete dropout. arXiv preprint arXiv:1705.07832, 2017.Google Scholar
- A. Graves. Practical variational inference for neural networks. In Advances in Neural Information Processing Systems, pages 2348--2356, 2011. Google ScholarDigital Library
- E. Gutmann, T. Pruitt, M. P. Clark, L. Brekke, J. R. Arnold, D. A. Raff, and R. M. Rasmussen. An intercomparison of statistical downscaling methods used for water resource assessments in the united states. Water Resources Research, 50(9):7167--7186, 2014.Google ScholarCross Ref
- M. Hessami, P. Gachon, T. B. Ouarda, and A. St-Hilaire. Automated regression-based statistical downscaling tool. Environmental Modelling & Software, 23(6):813--834, 2008. Google ScholarDigital Library
- B. Hewitson and R. Crane. Climate downscaling: techniques and application. Climate Research, pages 85--95, 1996.Google ScholarCross Ref
- H. Hidalgo, M. Dettinger, and D. Cayan. Downscaling with constructed analogues: Daily precipitation and temperature fields over the united states. 2008.Google Scholar
- G. E. Hinton and D. Van Camp. Keeping the neural networks simple by minimizing the description length of the weights. In Proceedings of the sixth annual conference on Computational learning theory, pages 5--13. ACM, 1993. Google ScholarDigital Library
- C. C. Jaeger, O. Renn, E. A. Rosa, and T. Webler. Decision analysis and rational action. Human choice and climate change, 3:141--216, 1998.Google Scholar
- R. W. Katz. Techniques for estimating uncertainty in climate change scenarios and impact studies. Climate Research, 20(2):167--185, 2002.Google ScholarCross Ref
- A. Kay, H. Davies, V. Bell, and R. Jones. Comparison of uncertainty sources for climate change impacts: flood frequency in england. Climatic Change, 92(1--2):41--63, 2009.Google Scholar
- A. Kendall, V. Badrinarayanan, and R. Cipolla. Bayesian segnet: Model uncertainty in deep convolutional encoder-decoder architectures for scene understanding. arXiv preprint arXiv:1511.02680, 2015.Google Scholar
- A. Kendall and Y. Gal. What uncertainties do we need in bayesian deep learning for computer vision? In Advances in Neural Information Processing Systems, 2017.Google ScholarDigital Library
- M. S. Khan, P. Coulibaly, and Y. Dibike. Uncertainty analysis of statistical downscaling methods. Journal of Hydrology, 319(1):357--382, 2006.Google ScholarCross Ref
- J. Kim, J. Kwon Lee, and K. Mu Lee. Accurate image super-resolution using very deep convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1646--1654, 2016.Google ScholarCross Ref
- D. Kingma and J. Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.Google Scholar
- C. Ledig, L. Theis, F. Huszár, J. Caballero, A. Cunningham, A. Acosta, A. Aitken, A. Tejani, J. Totz, Z. Wang, et al. Photo-realistic single image super-resolution using a generative adversarial network. arXiv preprint, 2016.Google Scholar
- D. B. Lobell, M. B. Burke, C. Tebaldi, M. D. Mastrandrea, W. P. Falcon, and R. L. Naylor. Prioritizing climate change adaptation needs for food security in 2030. Science, 319(5863):607--610, 2008.Google ScholarCross Ref
- Y. Lv, Y. Duan, W. Kang, Z. Li, and F.-Y. Wang. Traffic flow prediction with big data: a deep learning approach. IEEE Transactions on Intelligent Transportation Systems, 16(2):865--873, 2015.Google ScholarDigital Library
- X. Ma, H. Yu, Y. Wang, and Y. Wang. Large-scale transportation network congestion evolution prediction using deep learning theory. PloS one, 10(3):e0119044, 2015.Google ScholarCross Ref
- D. J. MacKay. A practical bayesian framework for backpropagation networks. Neural computation, 4(3):448--472, 1992. Google ScholarDigital Library
- D. Maraun, F. Wetterhall, A. Ireson, R. Chandler, E. Kendon, M. Widmann, S. Brienen, H. Rust, T. Sauter, M. Themeßl, et al. Precipitation downscaling under climate change: Recent developments to bridge the gap between dynamical models and the end user. Reviews of Geophysics, 48(3), 2010.Google ScholarCross Ref
- T. P. Miyanawala and R. K. Jaiman. An efficient deep learning technique for the navier-stokes equations: Application to unsteady wake flow dynamics. arXiv preprint arXiv:1710.09099, 2017.Google Scholar
- E. Racah, C. Beckham, T. Maharaj, S. Kahou, M. Prabhat, and C. Pal. Extremeweather: A large-scale climate dataset for semi-supervised detection, localization, and understanding of extreme weather events. In Advances in Neural Information Processing Systems, pages 3405--3416, 2017.Google Scholar
- M. A. Semenov and P. Stratonovitch. Use of multi-model ensembles from global climate models for assessment of climate change impacts. Climate research (Open Access for articles 4 years old and older), 41(1):1, 2010.Google Scholar
- J. M. L. Sloughter, A. E. Raftery, T. Gneiting, and C. Fraley. Probabilistic quantitative precipitation forecasting using bayesian model averaging. Monthly Weather Review, 135(9):3209--3220, 2007.Google ScholarCross Ref
- J. W. Taylor. A quantile regression neural network approach to estimating the conditional density of multiperiod returns. Journal of Forecasting, 19(4):299--311, 2000.Google ScholarCross Ref
- K. E. Taylor, R. J. Stouffer, and G. A. Meehl. An overview of cmip5 and the experiment design. Bulletin of the American Meteorological Society, 93(4):485--498, 2012.Google ScholarCross Ref
- T. Vandal, E. Kodra, and A. R. Ganguly. Intercomparison of machine learning methods for statistical downscaling: The case of daily and extreme precipitation. arXiv preprint arXiv:1702.04018, 2017.Google Scholar
- T. Vandal, E. Kodra, S. Ganguly, A. Michaelis, R. Nemani, and A. R. Ganguly. Deepsd: Generating high resolution climate change projections through single image super-resolution. In 23rd ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2017. Google ScholarDigital Library
- G. Wang, W. Li, M. A. Zuluaga, R. Pratt, P. A. Patel, M. Aertsen, T. Doel, A. L. David, J. Deprest, S. Ourselin, et al. Interactive medical image segmentation using deep learning with image-specific fine-tuning. IEEE Transactions on Medical Imaging, 2018.Google ScholarCross Ref
- D. Wuebbles, D. Fahey, K. Hibbard, D. Dokken, B. Stewart, and T. Maycock. Climate science special report: Fourth national climate assessment, volume i. 2017.Google Scholar
- Y. Zhu and N. Zabaras. Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification. arXiv preprint arXiv:1801.06879, 2018. Google ScholarDigital Library
Index Terms
- Quantifying Uncertainty in Discrete-Continuous and Skewed Data with Bayesian Deep Learning
Recommendations
Numerical approach for quantification of epistemic uncertainty
In the field of uncertainty quantification, uncertainty in the governing equations may assume two forms: aleatory uncertainty and epistemic uncertainty. Aleatory uncertainty can be characterised by known probability distributions whilst epistemic ...
A general framework for quantifying aleatoric and epistemic uncertainty in graph neural networks
AbstractGraph Neural Networks (GNN) provide a powerful framework that elegantly integrates Graph theory with Machine learning for modeling and analysis of networked data. We consider the problem of quantifying the uncertainty in predictions of ...
Uncertainty in Bayesian deep label distribution learning
AbstractThough widely used, most deep convolutional neural networks fail to capture prediction uncertainty, which can be crucial in scenarios such as automotive applications and disease diagnosis. Aleatoric and epistemic uncertainties have ...
Highlights- We define semantic uncertainty, which complements the aleatoric and model uncertainties, to evaluate the uncertainty related to data incompleteness and ...
Comments